Mike will be speaking about RDM and Master Data Management at Collaborate 2011.
Overview
This white paper focuses on the principals of Master Data Management (MDM) within the context of an overall Data Governance Strategy. It qualifies the value proposition MDM introduces while empowering organizations with the agility to be reactive in today’s economy. It dives into the principals of MDM using the framework of Oracle’s Data Relationship Manager (DRM) tool as a backdrop. In so doing, terms such as Versions, Hierarchies, Nodes, Properties, Validations, Verifications, and Exports are discussed.
Master Data Management Background
A key component to an organization’s Data Governance and Stewardship Strategy is Master Data Management (MDM). Above all other information management disciplines, MDM promotes one of the highest value-add benefits. Consequently, is it also one of the most resource intensive to implement. MDM spans enterprise jurisdictions, political fiefdoms, organizational hierarchies, technical architectures, business processes, and applications. No wonder the institution of an MDM strategy is as much about people and process as it is about technology. And, despite their best efforts, IT often struggles with realizing the promise of MDM while at the same time not crippling the business with bureaucracy and inflexibility. Assisting with this dilemma are technologies such as Oracle’s Data Relationship Manager (DRM). And, although data governance is not addressed by technology alone, DRM offers a foundation for establishing the disciplines, processes, and governance required to administer an Enterprise MDM Strategy.
What Is Master Data Management
Master Data Management (MDM) encompasses the people, processes, standards, and technology for ensuring the consistency and use of data throughout the enterprise and across business processes. Said another way - “The ability to analyze data across business functions and business processes is dependent upon linking data from disparate systems, applications and external sources. This ability to effectively link data from disparate sources is dependent upon master data.” – Jonathan Wu, DM Review
When we breakdown the information administered by MDM principals we find the following metadata components:
- Standard Naming Conventions
- Data Definitions
- Dimensions and/or Hierarchical Relationships
- Properties
- Mappings
- Verifications/Validations
- Versioning
Standard Naming Conventions. Naming standard for a master data element (node). In most cases long and short names are represented.
Data Definitions. Technical and Business description of the node being represented.
Dimension and/or Hierarchical Relationships. Represents the relationship one node has with others within its context (Dimension). This is usually denoted in a hierarchical fashion.
Property. Attributes that describe the nature of the node. Properties are organized into categories and can be leveraged during data extraction, mapping, verifications, and validations.
Mappings. Describes the relationship a node has in its context (Dimension/Hierarchy) with node(s) in alternative Dimensions/Hierarchies.
Verifications/Validation. Pre-defined rules that are executed upon altering the state of a node or performed in batch.
Versioning. Records the history and lineage of changes made to the master data records.
The Current Challenge
Part of the problem is the lack of understanding as to the deliverables from an MDM Strategy, and more importantly articulating the positive affect they have on the business’ capability to execute. In order to address this, one must first appreciate that most medium and all large organizations support multiple systems/application in order to conduct business. A bank for example may have separate credit card systems from their deposit systems, and yet others for processing mortgages, and still more for servicing Home Equity Lines. Even with the advent of the Enterprise Resource Planning (ERP) system, organizations must contend with capturing, storing, moving, integrating, and reporting information from multiple disparate systems.
This challenge is compounded by the fact that many of these systems are owned by siloed business units, with different leaders, often with individualistic and in the worst case conflicting charges. Additionally, the owners of these solutions are not rewarded for maintaining consistency across the organization and business processes. In fact, they are often encouraged to “make an exception” and bypass policies in the interest of “getting the business what they need ASAP; no matter the cost. We’ll fix it later…” – sound familiar?
This is not to suggest that exception handling is not are reality of doing business. In truth, every MDM Strategy must have a process for dealing with exceptions to the governance process. Unfortunately without a well vetted MDM process, exceptions become the norm and their employment becomes “just another day at the office”. Here in lies the problem. An exception should be a BIG DEAL. One in which the proverbial alarms sound up and down the corporate corridors. An extension of this occurrence is an “override” of the MDM processes administered by Oracle’s DRM solution.
Oracle’s Data Relationship Manager (DRM) as a Solution
Although master data challenges are not addressed by simply throwing technology at the problem, organizations are turning to software solutions to serve as the foundation for the establishment of the standards, disciplines, and governances required to maintain master data constructs. Oracle’s Data Relationship Manager (DRM) helps organizations solve the costly, labor-intensive problem of administering master data that is otherwise managed through phone calls, spreadsheets, and e-mails. Ultimately, Oracle DRM empowers the Information Worker by enabling them to become a vital/integral part of the master data management process.
Key benefits introduced by Oracle’s DRM solution consist of:
- Empowers business users to manage master data
- Improves change management processes and communications
- Ensures reporting integrity
- Controls and secures master data across the enterprise
- Establishes audit controls and accountability
- Centralizes, automates, and simplifies complex business rule management
- Delivers master data publishing, viewing, and downloading capabilities
- Accelerates financial reporting and eliminate redundancies
- Supports Versioning (pre-merger, reorganization, etc.)
- Promotes validation & verification rules
- Enforces balanced hierarchy (if required)
- Allows business users to become direct contributors to the iterative process of managing complex, rapidly changing metadata
In short, DRM’s value lies in systematically defining a set of governances that allow for the dissemination of master data management outside of IT and push it into the hands of business owners. Most of us in IT would suggest this proposition is “scary” at best and “suicide” at worst. Why – Because the business does not always have a clear understanding of the implications associated with master data changes and the systems for which this information is subscribed. But, with DRM the safeguards are systematically represented. For example, a Knowledge Worker is not allowed to add a new account to the Oracle R12 GL Chart of Accounts (COA) master record unless they provide representation for every Segment Field and map the new account to a corresponding node in the Hyperion Planning COA master record.
Master Data Management – The Big Picture
Serving as the foundation of an organization’s Master Data Management Strategy, Oracle’s DRM solution has multiple components. Each concept has relevance with regards to enforcing governance and oversight in the management of master data. Some key terminologies are…
- Dimensions, Hierarchies and Nodes: The master data “skeleton”. Provides the context for which each master record (node) is defined and correlated with other nodes.
- Versions: Identifies current versus prior manifestations of master records. Tracks history and lineage of changes.
- Properties and Property Categories: Qualifies and organizes metadata attributes that describe each node.
- Validations and Verifications: Rules that enforce data standards, process alignment, and system integration.
- Node Types: Special property that qualifies the disposition of the node from a business use perspective. i.e. Account Node, Entity Node, Product Node, etc.
- Security and Node Access Groups: Security provisioning that defines Group based access privileges to the master data
- Hierarchy Compares: Process by which one master data structure (Dimension) is compared with another whereby difference are reported.
- Imports: Facility for importing master record layouts in mass.
- Exports: Facility for publishing master data to subscribing systems.
As mentioned earlier, Oracle’s DRM solution is an excellent foundation for administering the disciplines necessary to support a Master Data Strategy, but this is only a component of the equation. People, process, and standards are equally important.
Some guiding principles to consider when deploying Master Data Management principals consist of:
Process before technology. Business processes rather than technology are the driving factor to MDM solutions.
Business ownership. Solution should be owned and driven by the business with IT playing a supporting role in ensuring that best practice and standards are adhered to.
Incremental deployment. Do not underestimate the time and expertise needed to develop a “foundational” MDM Strategy. A transformational approach is warranted. With the Goal State in mind, limit the scope of the initial deployment (i.e. Product Master Data).
Consider process impacts. Recognize that the application of MDM controls will bring up-front process performance penalties the dividends of which are realized strategically downstream.
Institute data governance policies and processes. Allow time and funding for people and process change management. The “politics” of MDM (Data Governance) will pose a challenge. For example - Finance may see the value of centralizing product master data management, but other factions may be reluctant to give up data ownership and control.
System impact analysis. Re-engineering Master Data structures will have an impact on critical processes and subscribing systems. Plan on a transition strategy that allows for scheduled and on-demand metadata synchronization.
Oracle DRM – Master Data Management Catalyst
When approaching Master Data Management initiatives using Oracle’s DRM technology, the solution is represented by four (4) Layers:
- Data Management Layer – Naming Standards, Hierarchies, Dimensions, Properties, Version Control, Mappings
- Business Logic Layer – Business Rules/Validations, Node Types
- System Interface Layer – Data Imports, Data Exports, Compares
- Governance – Security, Administration, Processes
As is so often the case with Master Data initiatives, IT and the Business often do not know what they want from the solution until it is shown to them (“playback”). The DRM utility is intended not only as the foundation for the Master Data Strategy, but as a tool for supporting the iterative nature of requirements gathering, design, and development. Within the context of each Layer, DRM promotes an iterative, graphical interface that facilitates the capture, integration, and playback of the business and technical requirements.
Is it conceivable to institute a Master Data Management Strategy in the absence of technology such as Oracle DRM? Yes. But, recognize that adoption is one of the largest hurdles to overcome with MDM initiative. Leveraging a tool that intuitively facilitates the capture and presentation of standards and rules for which master data will be governed is an enormous benefit. Also, realize that if the MDM Strategy includes the dissemination of master data control outside of IT, this vision will never be fully realized in the absence of the systemic controls and validations that can only be enforced with technology.
In the end, there needs to be an appreciation that the institution of MDM principals is a journey for which there is never truly a destination. As technical, business, economic, policy, and compliance mandates change, so will the need for your MDM solution to adapt. Using technology such as Oracle’s DRM solution reduces project risk, accelerate MDM initiatives, and extend master data management capabilities into the enterprise